Microeconometric Modeling William Greene Stern School of Business New York University New York NY USA 2.3 Panel Data Models for Binary Choice

Concepts Models Unbalanced Panel Attrition Bias Inverse Probability Weight Heterogeneity Population Averaged Model Clustering Pooled Model Quadrature Maximum Simulated Likelihood Conditional Estimator Incidental Parameters Problem Partial Effects Bias Correction Mundlak Specification Variable Addition Test Random Effects Progit Fixed Effects Probit Fixed Effects Logit Dynamic Probit Mundlak Formulation Correlated Random Effects Model

Application: Health Care Panel Data German Health Care Usage Data Data downloaded from Journal of Applied Econometrics Archive. This is an unbalanced panel with 7,293 individuals. They can be used for regression, count models, binary choice, ordered choice, and bivariate binary choice.  There are altogether 27,326 observations.  The number of observations ranges from 1 to 7.  (Frequencies are: 1=1525, 2=2158, 3=825, 4=926, 5=1051, 6=1000, 7=987).  Variables in the file are DOCTOR = 1(Number of doctor visits > 0) HOSPITAL = 1(Number of hospital visits > 0) HSAT =  health satisfaction, coded 0 (low) - 10 (high)   DOCVIS =  number of doctor visits in last three months HOSPVIS =  number of hospital visits in last calendar year PUBLIC =  insured in public health insurance = 1; otherwise = 0 ADDON =  insured by add-on insurance = 1; otherswise = 0 HHNINC =  household nominal monthly net income in German marks / 10000. (4 observations with income=0 were dropped) HHKIDS = children under age 16 in the household = 1; otherwise = 0 EDUC =  years of schooling AGE = age in years MARRIED = marital status 4

Unbalanced Panels Group Sizes Most theoretical results are for balanced panels. Most real world panels are unbalanced. Often the gaps are caused by attrition. The major question is whether the gaps are ‘missing completely at random.’ If not, the observation mechanism is endogenous, and at least some methods will produce questionable results. Researchers rarely have any reason to treat the data as nonrandomly sampled. (This is good news.) Group Sizes

Unbalanced Panels and Attrition ‘Bias’ Test for ‘attrition bias.’ (Verbeek and Nijman, Testing for Selectivity Bias in Panel Data Models, International Economic Review, 1992, 33, 681-703. Variable addition test using covariates of presence in the panel Nonconstructive – what to do next? Do something about attrition bias. (Wooldridge, Inverse Probability Weighted M-Estimators for Sample Stratification and Attrition, Portuguese Economic Journal, 2002, 1: 117-139) Stringent assumptions about the process Model based on probability of being present in each wave of the panel

Panel Data Binary Choice Models Random Utility Model for Binary Choice Uit =  + ’xit + it + Person i specific effect Fixed effects using “dummy” variables Uit = i + ’xit + it Random effects using omitted heterogeneity Uit =  + ’xit + it + ui Same outcome mechanism: Yit = 1[Uit > 0]

Ignoring Unobserved Heterogeneity

Ignoring Heterogeneity in the RE Model Scale factor is too large  is too small

Population average coefficient is off by 43% but the partial effect is off by 6%

Ignoring Heterogeneity (Broadly) Presence will generally make parameter estimates look smaller than they would otherwise. Ignoring heterogeneity will definitely distort standard errors. Partial effects based on the parametric model may not be affected very much. Is the pooled estimator ‘robust?’ Less so than in the linear model case.

Pooled vs. RE Panel Estimator ---------------------------------------------------------------------- Binomial Probit Model Dependent variable DOCTOR --------+------------------------------------------------------------- Variable| Coefficient Standard Error b/St.Er. P[|Z|>z] Mean of X Constant| .02159 .05307 .407 .6842 AGE| .01532*** .00071 21.695 .0000 43.5257 EDUC| -.02793*** .00348 -8.023 .0000 11.3206 HHNINC| -.10204** .04544 -2.246 .0247 .35208 Unbalanced panel has 7293 individuals Constant| -.11819 .09280 -1.273 .2028 AGE| .02232*** .00123 18.145 .0000 43.5257 EDUC| -.03307*** .00627 -5.276 .0000 11.3206 HHNINC| .00660 .06587 .100 .9202 .35208 Rho| .44990*** .01020 44.101 .0000

Partial Effects ---------------------------------------------------------------------- Partial derivatives of E[y] = F[*] with respect to the vector of characteristics They are computed at the means of the Xs Observations used for means are All Obs. --------+------------------------------------------------------------- Variable| Coefficient Standard Error b/St.Er. P[|Z|>z] Elasticity |Pooled AGE| .00578*** .00027 21.720 .0000 .39801 EDUC| -.01053*** .00131 -8.024 .0000 -.18870 HHNINC| -.03847** .01713 -2.246 .0247 -.02144 |Based on the panel data estimator AGE| .00620*** .00034 18.375 .0000 .42181 EDUC| -.00918*** .00174 -5.282 .0000 -.16256 HHNINC| .00183 .01829 .100 .9202 .00101

The Effect of Clustering Yit must be correlated with Yis across periods Pooled estimator ignores correlation Broadly, yit = E[yit|xit] + wit, E[yit|xit] = Prob(yit = 1|xit) wit is correlated across periods Assuming the marginal probability is the same, the pooled estimator is consistent. (We just saw that it might not be.) Ignoring the correlation across periods generally leads to underestimating standard errors.

‘Cluster’ Corrected Covariance Matrix

Cluster Correction: Doctor ---------------------------------------------------------------------- Binomial Probit Model Dependent variable DOCTOR Log likelihood function -17457.21899 --------+------------------------------------------------------------- Variable| Coefficient Standard Error b/St.Er. P[|Z|>z] Mean of X | Conventional Standard Errors Constant| -.25597*** .05481 -4.670 .0000 AGE| .01469*** .00071 20.686 .0000 43.5257 EDUC| -.01523*** .00355 -4.289 .0000 11.3206 HHNINC| -.10914** .04569 -2.389 .0169 .35208 FEMALE| .35209*** .01598 22.027 .0000 .47877 | Corrected Standard Errors Constant| -.25597*** .07744 -3.305 .0009 AGE| .01469*** .00098 15.065 .0000 43.5257 EDUC| -.01523*** .00504 -3.023 .0025 11.3206 HHNINC| -.10914* .05645 -1.933 .0532 .35208 FEMALE| .35209*** .02290 15.372 .0000 .47877

Fixed Effects Modeling Advantages: Allows correlation of covariates and heterogeneity Disadvantages: Complications of computing all those dummy variable coefficients – solved problem (Greene, 2004) No time invariant variables – not solvable in the FE context Incidental Parameters problem – persistent small T bias, does not go away Strategies Unconditional estimation Conditional estimation – Rasch/Chamberlain Hybrid conditional/unconditional estimation Bias corrrections Mundlak estimator

Fixed Effects Models Estimate with dummy variable coefficients Uit = i + ’xit + it Can be done by “brute force” for 10,000s of individuals F(.) = appropriate probability for the observed outcome Compute  and i for i=1,…,N (may be large) See FixedEffects.pdf in course materials.

Unconditional Estimation Maximize the whole log likelihood Difficult! Many (thousands) of parameters. Feasible – NLOGIT (2004) (“Brute force”)

Conditional Estimation Principle: f(yi1,yi2,… | some statistic) is free of the fixed effects for some models. Maximize the conditional log likelihood, given the statistic. Can estimate β without having to estimate αi. Only feasible for the logit model. (Poisson and a few other continuous variable models. No other discrete choice models.)

Binary Logit Conditional Probabilities

Example: Two Period Binary Logit

Example: Seven Period Binary Logit

The amount of computation increases very rapidly as T increases, but with modern computers, it is not a major constraint. Here is an example with T = 50. The data are generated from a probit process with b1 = b2 = .5. But, it is fit as a logit model. The coefficients obey the familiar relationship, 1.6*probit.

Estimating Partial Effects “The fixed effects logit estimator of  immediately gives us the effect of each element of xi on the log-odds ratio… Unfortunately, we cannot estimate the partial effects… unless we plug in a value for αi. Because the distribution of αi is unrestricted – in particular, E[αi] is not necessarily zero – it is hard to know what to plug in for αi. In addition, we cannot estimate average partial effects, as doing so would require finding E[Λ(xit + αi)], a task that apparently requires specifying a distribution for αi.” (Wooldridge, 2010) 26

MLE for Logit Constant Terms

An Approximate Solution for E[i]

An Approximate Solution for i

Experimental results look promising Experimental results look promising. In a simulation based on the health care data, the approximation is highly correlated with the truth.

Fixed Effects Logit Health Model: Conditional vs. Unconditional

Incidental Parameters Problems: Conventional Wisdom General: The unconditional MLE is biased in samples with fixed T except in special cases such as linear or Poisson regression (even when the FEM is the right model). The conditional estimator (that bypasses estimation of αi) is consistent. Specific: Upward bias (experience with probit and logit) in estimators of 

A Mundlak Correction for the FE Model

Mundlak Correction

Random Effects

Quadrature – Butler and Moffitt (1982)

Quadrature Log Likelihood

Random Effects Model: Quadrature ---------------------------------------------------------------------- Random Effects Binary Probit Model Dependent variable DOCTOR Log likelihood function -16290.72192  Random Effects Restricted log likelihood -17701.08500  Pooled Chi squared [ 1 d.f.] 2820.72616 Estimation based on N = 27326, K = 5 Unbalanced panel has 7293 individuals --------+------------------------------------------------------------- Variable| Coefficient Standard Error b/St.Er. P[|Z|>z] Mean of X Constant| -.11819 .09280 -1.273 .2028 AGE| .02232*** .00123 18.145 .0000 43.5257 EDUC| -.03307*** .00627 -5.276 .0000 11.3206 HHNINC| .00660 .06587 .100 .9202 .35208 Rho| .44990*** .01020 44.101 .0000 |Pooled Estimates Constant| .02159 .05307 .407 .6842 AGE| .01532*** .00071 21.695 .0000 43.5257 EDUC| -.02793*** .00348 -8.023 .0000 11.3206 HHNINC| -.10204** .04544 -2.246 .0247 .35208

A Variable Addition Test for FE vs. RE The Wald statistic of 45.27922 and the likelihood ratio statistic of 40.280 are both far larger than the critical chi squared with 5 degrees of freedom, 11.07. This suggests that for these data, the fixed effects model is the preferred framework.

A Dynamic Model

Dynamic Probit Model: A Standard Approach

Simplified Dynamic Model

A Dynamic Model for Public Insurance Age Household Income Kids in the household Health Status Add initial value, lagged value, group means

Dynamic Common Effects Model

Inverse Probability Weighting